SIS: Lets talk about geospatial vs geodetic

Ok,
So we see IDS now making the push on geodetic over geospatial. The
"round earth" is more accurate because it takes in consideration that
the earth really is elliptical in its calculations.

What that translates to is that the queries using geodetic are more
computational intensive than spatial queries.

Now in certain industries, this isn't a bad thing.

But if the bulk of the LBS market is geared towards hand held devices,
the majority of them are maybe accurate enough to pin point your
position within 10 meters at best. (That's 30 or so feet) Most likely,
you'll get within 100 feet in practice.

So why not use geospatial which is more than accurate enough for the
mass market?

What is the targeted niche that IDS marketing is trying to reach?

Oh, the other big thing. geospatial is free with IDS. Geodetic isn't.
Is it worth the 50K?

-G

PS. And what sample data ships with IDS these days?

Ian Michael Gumby wrote:
> PS. And what sample data ships with IDS these days?
>
>

Nothing. Ludwig Pauli called up and threatened IBM under the Data
Protection Act.

--
Cheers,
Obnoxio the Clown

http://obotheclown.blogspot.com

Ian Michael Gumby wrote:
> Ok,
> So we see IDS now making the push on geodetic over geospatial. The
> "round earth" is more accurate because it takes in consideration that
> the earth really is elliptical in its calculations.
>
> What that translates to is that the queries using geodetic are more
> computational intensive than spatial queries.
>
> Now in certain industries, this isn't a bad thing.
>
> But if the bulk of the LBS market is geared towards hand held devices,
> the majority of them are maybe accurate enough to pin point your
> position within 10 meters at best. (That's 30 or so feet) Most likely,
> you'll get within 100 feet in practice.
>
> So why not use geospatial which is more than accurate enough for the
> mass market?
>
> What is the targeted niche that IDS marketing is trying to reach?
>
> Oh, the other big thing. geospatial is free with IDS. Geodetic isn't.
> Is it worth the 50K?
>
> -G
>
> PS. And what sample data ships with IDS these days?

For small distances the difference doesn't matter, but if you want to
calculate the traveltime from Seattle to Yokohama sailing with 21 knots,
or if you want to calculate when a satelite passes a certain point, you
will get very different results using spatial and geodetic.
Size matters.

Claus Samuelsen said:

> Size matters.

That's not what she said last night!

--
Bye now,
Obnoxio

http://obotheclown.blogspot.com/

Obnoxio The Clown wrote:
> Claus Samuelsen said:
>
>> Size matters.
>
> That's not what she said last night!
>
She meant your shoes...


--
Serge Rielau
DB2 Solutions Development
IBM Toronto Lab

On Aug 24, 8:38 am, Claus Samuelsen wrote:
> Ian Michael Gumby wrote:
> > Ok,
> > So we see IDS now making the push on geodetic over geospatial. The
> > "round earth" is more accurate because it takes in consideration that
> > the earth really is elliptical in its calculations.
>
> > What that translates to is that the queries using geodetic are more
> > computational intensive than spatial queries.
>
> > Now in certain industries, this isn't a bad thing.
>
> > But if the bulk of the LBS market is geared towards hand held devices,
> > the majority of them are maybe accurate enough to pin point your
> > position within 10 meters at best. (That's 30 or so feet) Most likely,
> > you'll get within 100 feet in practice.
>
> > So why not use geospatial which is more than accurate enough for the
> > mass market?
>
> > What is the targeted niche that IDS marketing is trying to reach?
>
> > Oh, the other big thing. geospatial is free with IDS. Geodetic isn't.
> > Is it worth the 50K?
>
> > -G
>
> > PS. And what sample data ships with IDS these days?
>
> For small distances the difference doesn't matter, but if you want to
> calculate the traveltime from Seattle to Yokohama sailing with 21 knots,
> or if you want to calculate when a satelite passes a certain point, you
> will get very different results using spatial and geodetic.
> Size matters.

Sigh.

Remember I said this: "
> But if the bulk of the LBS market is geared towards hand held devices,
> > the majority of them are maybe accurate enough to pin point your
> > position within 10 meters at best. (That's 30 or so feet) Most likely,
> > you'll get within 100 feet in practice.
"
Flight plans, and navigation are a bit different and are not what is
considered part of the "bulk" of the LBS market.

Without getting in to a lot of detail there are only a couple of areas
where having tight positioning accuracy is required.
(Agricutlure, Civil engineering, Large Scale Construction, potential
navigation (flight/shipping) )

But even there, the GPS isn't really geodedic, with the exception of
Large Scale construction.
(In shipping, if I know my position is X, and I know the harbor is
position Y, I don't care if my position is off by 100 meters. Heck the
nominal position Y of the harbor could be off by as much as 1/2 a
click (km) due to where they position it in the harbor.)

In large scale construction, you *do* need to know your position, and
track your position along a geodetic course, (including z), so you can
better estimate you cost of materials.

I'll use Agriculture as an example.
You want to use GPS to help you when you spray your fields. So you set
up a Trimble base station, let it sit for a minimum of 48-72 hours so
that it can get a good fix on its position. You then have a GPS system
that ties in with the satellites and the local base station. This
will give you your position within 3 cm. This is accurate enough to
allow large farm machinery to not do too much overlap/underlap. Here
your guidance accuracy is due to your base station. (Do we need to
talk about solar radiation?)

In the next generation aircraft positioning systems, one of the
designs is to have the aircrafts broadcast their own position. Now if
you think about it, the aircraft is large, moving fast, so that it
will not matter if your current position is off by a couple of meters.
The flight computers have already been doing geodedic calculations, so
your flight tracking system doesn't have to be that accurate since all
it is doing is tracking the broadcast position from the plane. Your
airports may provide a "reference" point for planning the flight
information, but the airport itself is much larger than a "point".
(But now we're getting in to too much detail)

And there is one other issue. Your database could be just a geospatial
database with the application itself supporting the geodetic
calculations.
(Which goes back to my original question regarding IBM's positioning
statement of chasing geodetic over geospatial)


Sorry Claus, its not size but distance. And in the GPS world, the
smaller, and greater accuracy the better.

I mean if you can get an atomic clock the size of a playing deck of
cards (and the state of the art is much smaller than that.) You could
build yourself a very reliable and accurate base station. (Ok maybe
you wouldn't. But I think Marty Lurie ?sp? (Sorry Marty) would and if
I didn't have a wife who's spending me in to the poor house, I'd do
it.

But hey! What do I know? I don't wear a brass rat. ;-)

Later.

PS. For the record, I'd prefer geodetic because, well, I'm a nerd and
I want extreme accuracy in my positioning. As Moore's law continues,
you will see smaller and more accurate handhelds with lower costs in
the next 5-10 years if not sooner. Unfortunately your maps will only
be so accurate.